Glutamine prevents acute kidney injury by modulating oxidative stress and apoptosis in tubular epithelial cells

Abstract

Acute kidney injury (AKI) represents a common complication in critically ill patients that is associated with increased morbidity and mortality. In a murine AKI model induced by ischemia/reperfusion injury (IRI), we show that glutamine significantly decreases kidney damage and improves kidney function. We demonstrate that glutamine causes transcriptomic and proteomic reprogramming in murine renal tubular epithelial cells (TECs), resulting in decreased epithelial apoptosis, decreased neutrophil recruitment, and improved mitochondrial functionality and respiration provoked by an ameliorated oxidative phosphorylation. We identify the proteins glutamine gamma glutamyltransferase 2 (Tgm2) and apoptosis signal-regulating kinase (Ask1) as the major targets of glutamine in apoptotic signaling. Furthermore, the direct modulation of the Tgm2-HSP70 signalosome and reduced Ask1 activation resulted in decreased JNK activation, leading to diminished mitochondrial intrinsic apoptosis in TECs. Glutamine administration attenuated kidney damage in vivo during AKI and TEC viability in vitro under inflammatory or hypoxic conditions.

Keywords: Apoptosis; Cellular immune response; Immunology; Mitochondria; Nephrology.

Figure 1. Glutamine administration attenuates kidney damage and improves kidney function during IRI-induced AKI.

WT mice were subjected to sham or IRI surgery and received glutamine or saline 4 hours after reperfusion. The levels of urinary protein biomarkers specific for different parts of the kidney nephron were assessed by ELISA for all time points. NGAL (A, n = 5) represents the distal tubule function; KIM-1 (B, n = 5) represents the functionality of proximal tubules. TIMP-2 and IGFBP7 (C, n = 5) are biomarkers of G₁ cell cycle arrest correlating with renal tubular cell stress. Plasma creatinine (D, n = 5) as well as blood urea nitrogen levels were measured (E, n = 5). Creatinine clearance was calculated 12 and 24 hours after the procedure (F, n = 5). Renal blood flow was analyzed in the Arteria renalis (A. renalis) at baseline conditions, 4 hours and 24 hours after IRI induction (G, n = 3; mean + SD). Neutrophil recruitment into the kidney was analyzed by flow cytometry (H, n = 5). After performance of H&E staining of paraffin-embedded sections 1 tissue section per mouse was scored to identify tissue damage (I, tubular injury score; J, representative H&E staining, n = 5). Black scale bar: 100 μm; white scale bar: 50 μm. The plasma levels of CXCL1 (K), CXCL2 (L), TNF-α (M), and IL-10 (N) were analyzed by ELISA (n = 5; mean ± SEM; 1-way ANOVA *P < 0.05; **P < 0.005; ***P < 0.001).

Figure 2. Glutamine affects reduction-oxidation capacity, NAD metabolism, and apoptotic processes.

WT mice were subjected to sham or IRI surgery and received glutamine or saline 15 minutes after reperfusion. Kidneys were collected and homogenized 24 hours after IRI induction. Mass spectrometric label-free quantification was performed in order to identify alteration in protein expression levels as a result of glutamine treatment. A hierarchical clustering heatmap indicates differentially expressed genes (rows) between the respective sham and IRI groups (A, n = 4). Further analysis of 4 kidney homogenates of glutamine-treated mice and saline-treated mice after IRI induction revealed the effect of glutamine on renal protein expression. Red indicates upregulation and blue indicates downregulation (B, n = 4). Volcano plots generated to compare glutamine versus saline treatment after IRI induction reveal 190 proteins to be significantly differentially expressed (C, n = 4). Among these proteins, 96 proteins were significantly elevated due to glutamine treatment (indicated by red dots), whereas 94 proteins were significantly decreased (indicated by blue dots). GO terms representing molecular function are presented in red, cellular component in green, and biological processes in blue (D; Fisher’s exact test, P value < 0.05; only significantly modulated GO terms are displayed). (E) Schematic illustration of signaling pathways in renal TECs affected by glutamine treatment.

Figure 3. Glutamine administration functionally attenuates renal tubular cell apoptosis.

WT mice were subjected to sham or IRI surgery and received glutamine or saline. Paraffin-embedded tissue sections were prepared and TUNEL staining was performed (A, TUNEL-positive cells [%]; B, representative TUNEL-stained cortex and medulla tissue) (n = 4; scale bar: 50 μm). Caspase-3 activity was detected in kidney lysates (C, n = 4). Western blotting was performed to assess the expression level of HSP70 (D and E), p-JNK (D and F), 14-3-3ζ and p–14‑3‑3ζ (Thr232) (D and G), Bad and p-Bad (Ser136) (D and H), caspase-3 (D and I), as well as apoptosis signal-regulating kinase (Ask1) and p-Ask1 (D and J). TECs were transfected with Ask1 siRNA and control siRNA using the Lipofectamine RNAiMAX Reagent. Knockdown efficiency determined by Western blot analysis was ~10% protein expression. Transfected TECs were treated with glutamine or saline, then subjected to hypoxia. TEC lysates were analyzed by Western blotting to assess the expression levels of HSP70 (K and L, n = 3) and caspase-3 (K and M, n = 3). MALDI imaging mass spectrometry (MALDI-IMS) of kidney sections was performed to analyze protein distribution of HSP70, 14-3-3ζ, and Bad (N). The scale represents the relative intensity of the protein (m/z, mass-to-charge ratio). Mean ± SEM, 1-way ANOVA *P < 0.05; **P < 0.005; ***P < 0.001.

Figure 4. Glutamine improves kidney function upon IRI by modulating Tgm2-HSP70 interaction.

WT mice were subjected to sham or IRI surgery and received glutamine or saline. Western blotting was performed to assess the expression of Tgm2 levels (A and B). MALDI-IMS of kidney sections was performed in order to analyze Tgm2 distribution (C). In addition to glutamine or saline exposure, mice obtained Tgm2 inhibitor ERW1041E or DMSO as vehicle control by an intravenous injection (D–F). Plasma creatinine levels (D, n = 5) and neutrophil recruitment into the kidney (E, n = 5) as well as Tgm2 activity in kidneys (F, n = 4) were determined 12 hours after IRI induction. WT and conditional Tgm2-KO mice (Tgm2^fl/fl sGLT2^cre+) were subjected to IRI surgery. Plasma (G, n = 3) and urine creatinine (H, n = 3) as well as neutrophil recruitment (I, n = 3) and the renal tubular injury score (J, n = 3) were assessed 24 hours after IRI. Mean ± SEM, 1-way ANOVA *P < 0.05; **P < 0.005; ***P < 0.001.

Figure 5. Glutamine supports mitochondrial respiration during oxidative stress.

WT mice were subjected to sham or IRI surgery and received glutamine or saline. NAD⁺/NADH ratio (A, n = 5), ATP concentration (B, n = 3), and NO₂^– concentration (C, n = 4) were detected in kidney suspensions 24 hours after IRI by colorimetric assays. TECs were treated with glutamine or saline and subsequently subjected to hypoxic (1% O₂) conditions for 24 hours. The mitochondrial membrane potential was assessed by TMRE fluorescence detection (D, n = 3). Mitochondrial ROS production was detected to assess oxidative stress (E, n = 3). Mitochondrial respiration was assessed using the Seahorse XF24 Flux Analyzer. The oxygen consumption rate (OCR) to analyze mitochondrial respiration was measured using the Mito Stress Kit (F and G; n = 3). Mean ± SEM; 1-way ANOVA. *P < 0.05; **P < 0.005; ***P < 0.001. FCCP, carbonyl cyanide-p-trifluoromethoxyphenylhydrazone.

Figure 6. Glutamine affects the expression of apoptosis- and oxidative stress–related proteins in renal TECs on the transcriptional and proteomic level.

TECs were identified in kidney homogenates by FACS based on the cell-specific marker prominin-1. RNA was isolated and sequenced. Hierarchical clustering heatmaps of RNA-Seq indicate differentially expressed genes (rows) between TECs of glutamine-treated mice after IRI and glutamine treatment compared with vehicle control (A, n = 3). Blue color indicates downregulation and red color indicates upregulation. Volcano plots generated to compare glutamine versus saline treatment after IRI induction identify 481 differently expressed genes in renal TECs (B). GO pathway enrichment analyses specify the fold enrichment of the transcriptome sequencing relative to the Mus musculus genome in TECs (C). GO terms representing molecular function are presented in red, cellular component in green, and biological processes in blue (Fisher’s exact test, P value < 0.05). TECs were isolated from murine kidneys and exposed to TNF-α for 18 hours or hypoxia for 24 hours. Additionally, TECs were treated with glutamine or saline as well as Tgm2 inhibitor ERW1041E or vehicle control, respectively. Caspase-3 activity was detected after TNF-α (n = 5) or hypoxia stimulation (n = 3, D), and Tgm2 levels were determined by Western blotting (E and F). Tgm2 activity in kidneys was assessed (G, n = 4). Immuno- and coimmunoprecipitated proteins were separated by SDS-PAGE and blotted using the indicated antibodies. Whole-cell lysate (INPUT) was used as protein control (H, n = 3). Mean ± SEM; 1-way ANOVA *P < 0.05; **P < 0.005; ***P < 0.001.